MySQL Instance Manager is been deprecated in MySQL 5.1 and is removed in MySQL 5.5.
Instance Manager uses its standard configuration file unless it
is started with a
--defaults-file option that
specifies a different file. On Windows, the standard file is
my.ini in the directory where Instance
Manager is installed. On Unix, the standard file is
/etc/my.cnf. (Prior to MySQL 5.0.10, the
MySQL Instance Manager read the same configuration files as the
MySQL Server, including
~/.my.cnf, and so forth.)
Instance Manager reads options for itself from the
[manager] section of the configuration file,
and options for server instances from
[manager] section contains any
of the options listed in
Section 18.104.22.168, “MySQL Instance Manager Command Options”, except for
those specified as having to be given as the first option on the
command line. Here is a sample
# MySQL Instance Manager options section [manager] default-mysqld-path = /usr/local/mysql/libexec/mysqld socket=/tmp/manager.sock pid-file=/tmp/manager.pid password-file = /home/cps/.mysqlmanager.passwd monitoring-interval = 2 port = 1999 bind-address = 192.168.1.5
section specifies options given by Instance Manager to a server
instance at startup. These are mainly common MySQL Server
options (see Section 5.1.3, “Server Command Options”). In addition, a
can contain the options in the following list, which are
specific to Instance Manager. These options are interpreted by
Instance Manager itself; it does not pass them to the server
when it attempts to start that server.
The Instance Manager-specific options must not be used in a
[mysqld] section. If a server is started
without using Instance Manager, it will not recognize these
options and will fail to start properly.
The path name of the mysqld server binary to use for the server instance.
This option disables Instance Manager monitoring functionality for the server instance. By default, an instance is guarded: At Instance Manager start time, it starts the instance. It also monitors the instance status and attempts to restart it if it fails. At Instance Manager exit time, it stops the instance. None of these things happen for nonguarded instances.
The number of seconds Instance Manager should wait for the
server instance to shut down. The default value is 35
seconds. After the delay expires, Instance Manager assumes
that the instance is hanging and attempts to terminate it.
If you use
InnoDB with large tables, you
should increase this value.
Here are some sample instance sections:
[mysqld1] mysqld-path=/usr/local/mysql/libexec/mysqld socket=/tmp/mysql.sock port=3307 server_id=1 skip-stack-trace core-file skip-bdb log-bin log-error log=mylog log-slow-queries [mysqld2] nonguarded port=3308 server_id=2 mysqld-path= /home/cps/mysql/trees/mysql-5.0/sql/mysqld socket = /tmp/mysql.sock5 pid-file = /tmp/hostname.pid5 datadir= /home/cps/mysql_data/data_dir1 language=/home/cps/mysql/trees/mysql-5.0/sql/share/english log-bin log=/tmp/fordel.log